The Royal Free London (RFL) and DeepMind Health have worked in partnership to develop a mobile application, Streams, to transform the care of patients with acute kidney injury (AKI). Streams provides real time patient deterioration alerts to clinicians mobile devices ; ability to review relevant clinical data at the bedside ; access to a standardised care protocol ; and enables communication within and between clinical teams.
Throughout the project, the care team consisting of nephrologists and critical care outreach nurses have been fully engaged in the design, testing and live use of Streams. The real time push notification of deteriorating patients with AKI prevents delays in treatment during the golden hours and enables the standardised treatment of patients using a specially-designed care protocol.
An independent evaluation undertaken by UCL, which is due to be published in the coming weeks, found that the introduction of Streams led to improvements in the speed and reliability of detection of patient deterioration; improved patient outcomes; and enhanced clinician experience. Further details will be published in the service evaluation
In the UK, acute kidney injury (AKI) contributes to 20% of all hospital admissions and causes up to 100,000 deaths, of which 30% could be prevented if patients received the right care sooner. It is associated with a need for longer stays in hospital, an increased use of resources, and it costs the NHS £1.2 billion per annum.
The challenge was that existing pathways for the identification and communication of patient deterioration are often dependent on outdated technologies such as pagers which results in delays to the delivery of key treatments by hours or days.
The development of Streams addresses this by providing clinicians with real time push alerts for deteriorating AKI patients ; collates the right clinical context information from multiple disparate IT solutions to enable them to deliver key treatments quicker ; and supports clinicians decision making using a standardised care bundle, including the assessment of life threatening complications of AKI, sepsis and hypo-perfusion, toxicity, obstruction and primary renal disease.
The goals of the RFL-DeepMind Health partnership are to address the quadruple aim ; to improve clinical outcomes ; enhance patient experience ; enhance clinician experience ; and reduce the cost of care.
The UCL independent review findings, which are still subject to peer review, show a number of significant improvements in the speed and reliability of AKI recognition, as well as improvements in patient outcomes, clinician experience, and costs of care.
The Streams app has been implemented to manage patients at our specialist tertiary centre site at the Royal Free Hospital in Hampstead, north London. We are currently in the process of extending the use of the system to the other hospitals within our Trust (Barnet Hospital and Chase Farm Hospital).
This has required some variation in the personnel used to deploy Streams and the feedback suggests that the device is intuitive. We have demonstrated the app in a wide range of public events e.g. staff and patient events in our hospitals and at national and international conferences. Streams is also a key innovation and interoperability component of our Global Digital Exemplar (GDE), which will provide a high profile platform to promote our learning across the NHS and globally.
DeepMind Health are working with a number of other NHS Trusts to implement the Streams app. Other Trusts will benefit from the intellectual property developed by both the RFL and DeepMind in the development of the app and the standardised clinical protocols. For example, Imperial College Healthcare NHS Trust recently went live with Streams for general results viewing based on the technical development work undertaken by the RFL and DeepMind Health.
The digitally-enabled care pathway was designed in collaboration with clinicians who are responsible for the service delivery. Clinical user groups were involved in hundreds of hours of user testing of Streams, ensuring the app integrated seamlessly into existing workflows. In response to feedback from users, the app was developed to optimise usability and impact.
The design and conduct of the independent service evaluation involved extensive patient and public involvement and engagement, and the evaluation was subject to regular scrutiny by a committee of clinicians and academics from RFH and beyond, as well as patients.
Mary Emerson, lead resuscitation nurse said: “I love the Streams app, it’s the reason I’m still at the Royal Free Hospital. It’s the best thing to have happened - in terms of innovation – that has helped me do my job better.
“It’s been developed with clinicians in mind, so it works for us and if we wanted things changed, that’s been listened to and responded to. It’s intuitive, so it doesn’t need much training, and it’s mobile, so I can show patients what their blood results are doing. That’s meaningful to them and helps them understand their care and what’s happening to them.”